Dissertation
Topics in Health Econometrics
Doctor of Philosophy (PhD), Washington State University
01/2012
Handle:
https://hdl.handle.net/2376/4270
Abstract
This dissertation discusses three topics in health econometrics focusing on misclassification in binary data, sources and the nature of bias, the impact of using misclassified data in econometric estimations and methods to identify and correct the bias. The first chapter proposes an econometric estimator to estimate correct inferences when the dependent variable of the binary choice model is endogenously misclassified. The approach is validated using a simulation study and applied to the analysis of a treatment program designed to improve family dynamics. The second chapter analyses the `HIV-virgin puzzle' where a number of adolescents in Africa who were found HIV positive report as never having sex, which indicates either the dominance of non-sexual modes of HIV transmission or systematic misreporting of sexual behavior. A method is proposed to estimate the extent of misreporting and the contribution of sexual mode for HIV transmission in Africa after accounting for misreporting. The third chapter employs econometric techniques to compare self-reported and objectively measured smoking data taking into account errors with both methods and cautions that objective measures may not always be more reliable than self-reported data.
Metrics
5 File views/ downloads
20 Record Views
Details
- Title
- Topics in Health Econometrics
- Creators
- Vidhura Senani B W Tennekoon
- Contributors
- Robert E Rosenman (Advisor)Ron C Mittelhammer (Advisor)Laura G Hill (Committee Member)Bidisha Mandal (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Economic Sciences, School of
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Number of pages
- 97
- Identifiers
- 99900581454001842
- Language
- English
- Resource Type
- Dissertation